A numerical approach to stochastic reach-avoid problems for Markov Decision Processes.
Nikolaos KariotoglouMaryam KamgarpourTyler H. SummersJohn LygerosPublished in: CoRR (2014)
Keyphrases
- markov decision processes
- optimal policy
- stochastic shortest path
- planning under uncertainty
- state space
- decision processes
- decision problems
- reinforcement learning
- reachability analysis
- dynamic programming
- partially observable
- policy iteration
- state and action spaces
- decentralized control
- continuous state spaces
- finite state
- transition matrices
- factored mdps
- risk sensitive
- markov decision process
- probabilistic planning
- real time dynamic programming
- decision theoretic planning
- control policies
- data mining
- partially observable markov decision processes
- reinforcement learning algorithms